Surgical Robot Trajectory Modeling for Precise Torque Sizing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robotic systems for minimally-invasive surgery face challenges in accurately sizing motors, transmissions, and brakes due to variability in user input for robotic movements, leading to either oversized components resulting in increased size and cost or undersized components causing delayed responsiveness.
Innovation Solution
The method involves generating a synthetic or virtual trajectory based on user data to estimate dynamic torque, using relationships between velocity and acceleration, and optimizing parameters to fit high-coverage directions for each pose of the robot, allowing for more precise sizing of robotic components.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If motor, transmission, and brake components are oversized to accommodate maximum user input variability, then responsiveness and reliability are improved, but system size and cost increase
Solution Approach 1:
The patent transforms the trajectory representation from time-dependent to pose-dependent parameters. By representing trajectories as sequences of poses with associated velocity and acceleration vectors rather than time-stamped position data, the system can evaluate all possible movement directions at each pose independently. This parameter transformation enables determination of maximum dynamic torque requirements without relying on conservative oversizing, allowing components to be sized precisely for actual operational needs.
2Weight of stationary object
If motor, transmission, and brake components are downsized to reduce system size and cost, then efficiency and cost are improved, but responsiveness and reliability deteriorate
Solution Approach 1:
The patent performs preliminary evaluation of all possible velocity and acceleration vector combinations at each robot pose before finalizing component sizing. By pre-calculating the maximum dynamic torque requirements across the complete pose space with comprehensive vector sampling, the system identifies the true maximum loads without needing to oversize components. This preliminary analysis ensures components are sized exactly for the maximum expected loads, preventing both oversizing and undersizing.
3Measurement precision
If multiple user trajectory samples are collected to account for movement variability, then accuracy of sizing is improved, but complexity of data processing increases
Solution Approach 1:
The patent creates a synthetic virtual trajectory that copies and synthesizes the essential characteristics of multiple user trajectories into a single representative model. Instead of processing numerous individual user motion logs, the system generates a virtual trajectory that encompasses the full range of velocity and acceleration directions at each pose. This synthetic copy captures the variability of multiple users while simplifying analysis to a single comprehensive trajectory evaluation.
Solution Approach 2:
The patent introduces pose-dependent velocity and acceleration vectors as an intermediary representation between raw user trajectory data and component sizing calculations. By transforming diverse user input trajectories into a unified pose-space vector representation, the system creates an intermediary model that preserves all essential movement characteristics while enabling systematic evaluation of maximum torque requirements without processing complexity of individual user data.
4Reliability
If time-dependent trajectory data is used from user input, then realism of motion patterns is improved, but difficulty in segregating relevant situations increases
Solution Approach 1:
The patent inverts the conventional approach by representing trajectories as sequences of poses with associated velocity and acceleration vectors rather than as time-stamped position data. This inversion transforms the representation from temporal sequence to spatial configuration space, allowing all velocity and acceleration directions at each pose to be evaluated independently. The inversion enables systematic identification of maximum torque requirements without needing to interpret time-dependent user behavior patterns.
Data Source
AI summary
For kinetic sizing, the dynamic torque to be provided by a robotic system may be based off of, in part, a maximum acceleration. Rather than trying to extract maximum acceleration from many samples, a relationship of velocity to acceleration from repetitive user inputs relative to a non-surgical target in different situations (e.g., accurate, fast, or balance tracing of the target movement) is established. The velocity for any given situation may be used to estimate the acceleration from the relationship. Rather than using many trajectory samples from many users, a synthetic trajectory may be used. The synthetic trajectory may be fit to user data while maintaining high-coverage properties for direction of movement for any given pose of the robot. Alternatively, a virtual trajectory decoupled from time is used. The virtual trajectory samples the directions at any given pose in a global high-coverage manner, without specifically using a time-dependent sequence of poses.